Skip to content
Agentmart
Submit a tool

remy-ci

Analyze CI/CD failure logs to diagnose build, test, and gate failures.

Skill by MyPeacefulValentine

IndexedMIT★ 28
Write a review

Chalk Labs

Need help deploying remy-ci?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mypeacefulvalentine-remy-ci

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Analyze CI/CD failure logs to diagnose build, test, and gate failures. Supports GitHub Actions (gh CLI), local log files, and pasted logs. Produces evidence packets for /remy-patch.

Use cases: testing, devops

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/remy-ci

Tools it pre-approves 7

From allowed-tools in SKILL.md: the agent may use these without asking you each time.

  • Read,
  • Grep,
  • Glob,
  • Bash,
  • PowerShell,
  • Write,
  • AskUserQuestion

What's inside

Skill name
remy-ci
Description
Analyze CI/CD failure logs to diagnose build, test, and gate failures. Supports GitHub Actions (gh CLI), local log files, and pasted logs. Produces evidence packets for /remy-patch.
License
MIT
Allowed tools
Read,, Grep,, Glob,, Bash,, PowerShell,, Write,, AskUserQuestion
Instructions
263 lines in SKILL.md · view on GitHub

Before installing

  • Pre-approves some shell commands

Agent Skills spec issues

  • body: contains an absolute path from the author's machine

Files (6)

  • README.md
  • README_zh.md
  • SKILL.md
  • diagnosis_template.md
  • output_schema.json
  • parsers.json

SKILL.md (first lines)

# Remy CI Protocol

CI/CD failure log analysis skill. Parses build/test/gate failure logs, localizes errors to source files, and produces a structured diagnosis report plus an evidence packet compatible with `/remy-patch`. Does NOT modify source code.

## 0. Configuration

| Environment Variable | Default | Description |
| :--- | :--- | :--- |
| `CI_LOG_MAX_LINES` | `500` | Maximum lines to retain per failed step when truncating logs. |

### Argument Parsing

```
/remy-ci [run_id | log_file_path | --paste]
```

1. If argument is a numeric string: treat as **GitHub Actions run ID** (Mode C).
2. If argument is a file path that exists on disk: treat as **local log file** (Mode B).
3. If argument is `--paste`: treat as **paste mode** (Mode A) — prompt user to provide log text.
4. If no argument provided: enter **Guided Mode** (Phase 0.1).

---

## Phase 0: Log Acquisition

**Goal**: Obtain CI failure log content for analysis.

### 0.1 Guided Mode (no arguments)

Use `AskUserQuestion` to ask:
- Input source? Options:
  - "GitHub Actions (gh CLI)" → proceed to 0.2
  - "Local log file" → proceed to 0.3
  - "Paste log text" → proceed to 0.4

### 0.2 GitHub Actions Mode (Mode C)

**Prerequisite Check**:
1. Run `Bash("gh --version 2>/dev/null && echo GH_OK || echo GH_MISSING")`.
2. If `GH_MISSING`: report "gh CLI not available. Use Mode A (paste) or Mode B (file) instead." → return to 0.1.
3. Run `Bash("gh auth status 2>&1 | head -3")` to verify authentication.
4. If not authenticated: report authentication issue → return to 0.1.

**Log Retrieval**:

If run_id was provided:
1. Fetch structured metadata: `Bash("gh run view {run_id} --json jobs,conclusion,name,headBranch")`.
2. Parse JSON output. Identify failed jobs and steps.
3. Fetch failed step logs: `Bash("gh run view {run_id} --log-failed 2>&1 | tail -n {CI_LOG_MAX_LINES}")`.

If no run_id:
1. Detect current branch: `Bash("git branch --show-current")`.
2. Find latest failed run: `Bash("gh run list --branch {branch} --status failure -L 1 --json databaseId,name,conclusion,event")`.
3. If no failed runs found: report "No failed CI runs found on branch {branch}." → HALT.
4. Extract `databaseId` and proceed as if run_id was provided.

Record the structured metadata as **CI Metadata** and the log text as **Raw Log**.

### 0.3 File Mode (Mode B)

1. `Read` the specified file path.
2. If file exceeds `CI_LOG_MAX_LINES` lines: apply Phase 1 truncation strategy.
3. Record content as **Raw Log**. CI Metadata is unavailable.

### 0.4 Paste Mode (Mode A)

1. Use `AskUserQuestion`: "Paste the CI failure log below (or provide the key error messages)."
2. Record user input as **Raw Log**. CI Metadata is unavailable.

**Phase 0 Exit**: Raw Log exists (and optionally CI Metadata). Proceed to Phase 1.

---

## Phase 1: Classification & Truncation

**Goal**: Identify the failure type(s) and reduce log volume.

### 1.1 Structured Classification (Mode C only)

If CI Metadata is available:
1. Parse job/step names to infer failure categories (e.g., step named "Build" → compile_error, step named "Test" → test_failure).
2. Map each failed step to a failure type from `parsers.json`.
3. If mapping succeeds: skip keyword classification (1.2) for those steps.

### 1.2 Keyword Classification (All modes, fallback)

1. `Read` the file `skills/remy-ci/parsers.json`.
2. For each registered parser entry, scan the Raw Log for its `keywords` (case-insensitive).
3. Count keyword matches per type. Assign the type(s) with the most matches.
4. If no keywords match any type: assign `unknown`.

### 1.3 Log Truncation

If Raw Log exceeds `CI_LOG_MAX_LINES`:
1. Extract all lines matching error/failure keywords across all matched parser types.
2. Extract the last `CI_LOG_MAX_LINES / 2` lines of the Raw Log (tail context).
3. Merge (deduplicate) these two sets, preserving original line order.
4. Record the result as **Filtered Log**.

If Raw Log is within limits: use Raw Log directly as **Filtered Log**.

**Phase 1 Exit**: Failure type(s) identified. Filtered Log available. Proceed to Phase 2.

---

## Phase 2: Type-Specific Parsing

**Goal**: Extract structured error information from the Filtered Log.

For each identified failure type, apply the corresponding parsing rules:

### 2.1 Compile Error (`compile_error`)
- Pattern: `{file}:{line}:{col}: error: {message}` or `{file}:{line}: error: {message}`
- Extract: file path, line number, column (if present), error message.

### 2.2 Link Error (`link_error`)
- Pattern: `undefined reference to '{symbol}'`, `multiple definition of '{symbol}'`
- Extract: symbol name, referencing object file (if present).

From MyPeacefulValentine/Remy-CC, MIT.

Signals

Stars
28
Forks
0
Open issues
0
Last commit (this folder)
Oct 17, 2026 (1 month ago)
Last release
Oct 18, 2026 (1 month ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get remy-ci running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying remy-ci?

CallGet help
Cite this page

remy-ci skill. Agent Mart. https://agentmart.dev/skills/mypeacefulvalentine-remy-ci (last updated 2026-10-24)